International Journal of Computer Networks and Communications Security

Volume 5, Issue 10, October 2016

 

 

Home Automation Using Smart Watch
 

Home Automation Using Smart Watch

Pages: 223-226 (4) | [Full Text] PDF (414 KB)
U Varma, A.V.S.S.Bharadwaj
Department of Electronic and Communications, GITAM University, Visakhapatnam, Andhra Pradesh, 530013, India

Abstract -
The process of Home automation is now widespread these days which is used in numerous day to day applications in variety of industries. This paper proposes an efficient and a quick solution for controlling of various home appliances using a smart watch. This is implemented using a raspberry pi board which uses flask framework and python programming for developing a web application that can communicate with home automation app on the Pebble Smart watch. The home automation app for smart watch is developed using JavaScript that uses Pebble API. In this paper basic variables, ON state and OFF state of an appliance are displayed in the home automation app in the smart watch.
 
Index Terms - Raspberry Pi, Raspbian OS, cloud pebble SDK, Flask framework, Python, Java-script, Relay, Bluetooth

Citation - U Varma, A.V.S.S.Bharadwaj. "Home Automation Using Smart Watch." International Journal of Computer Science and Software Engineering 5, no. 10 (2016): 223-226.

Data Fusion in Patient Centered Health Information Retrieval
 

Data Fusion in Patient Centered Health Information Retrieval

Pages: 227-232 (6) | [Full Text] PDF (366 KB)
NP Motlogelwa, E Thuma, T Leburu-Dingalo
Department of Computer Science, University of Botswana, Gaborone, P/BAG UB 00704, Botswna

Abstract -
When faced with medical ailments, a majority of lay people often submit circumlocutory queries to modern web search engines for self-diagnosis. However, such queries often fail to retrieve relevant documents to answer their information need. In this articles, we attempt to improve the retrieval effectiveness of such systems by enriching these circumlocutory queries with the most informative terms from two different external collections to produce two different expanded queries. In addition, we submit these expanded queries to the local collection (collection being searched) to produce two different rankings. Furthermore, we deploy data fusion techniques to combine these multiple rankings in order to further improve the retrieval effectiveness of such systems. Our empirical evaluation shows marked improvement in the retrieval performance of such systems when we fuse these multiple rankings. In particular, we see an improvement in both precision at 5 (P@5), precision at 10 (P@10) and recall when the number of rankers deployed for each data fusion technique is increased.
 
Index Terms - Data Fusion, Collection Enrichment, Query Expansion

Citation - NP Motlogelwa, E Thuma, T Leburu-Dingalo. "Data Fusion in Patient Centered Health Information Retrieval." International Journal of Computer Science and Software Engineering 5, no. 10 (2016): 227-232.

Classification of Parkinsons Disease using MRI Images
 

Classification of Parkinsons Disease using MRI Images

Pages: 233-237 (5) | [Full Text] PDF (515 KB)
S Pazhanirajan, P Dhanalakshmi
Assistant Professor, Department of Computer Science and Engineering, Annamalai University, Chidambaram, Tamilnadu 608002, India
Associate Professor, Department of Computer Science and Engineering, Annamalai University, Chidambaram, Tamilnadu 608002, India

Abstract -
A novel method for automatic classification of magnetic resonance image (MRI) under categories of normal and Parkinsons disease (PD) is then classified according to the severity of the medical specialty drawbacks. In recent years, with the advancement in all fields, human suffers from numerous specialty disorders like brain disorder, epilepsy, Alzheimer, Parkinson, etc. Parkinsons involves the malfunction and death of significant nerve cells within the brain, known as neurons. As metal progresses, the quantity of Dopastat made within the brain decreases, defeat someone, and make them unable to manage movements commonly. In the planned system, T2 (spin-spin relaxation time)—weighted MR images are obtained from the potential PD subjects. For categorizing the MRI knowledge, histogram features and Gray Level Co-occurrence Matrix (GLCM) features are extracted. The features obtained are given as input to the two different classifier techniques namely Support Vector Machine (SVM) and Radial Basis Function Neural Network (RBFNN). The classifiers, classifies the categories into normal or abnormal PD. Abnormal PD is classified into three Parkinsons diseases. Two different classifiers are used to classify the three subcategories namely Mild, Moderate and Advanced.
 
Index Terms - Parkinsons Disease (PD), Region of Interest Cropping (ROI), Gray-level Co-occurrence Matrix (GLCM), Support Vector Machine (SVM), Radial Basis Function Neural Network (RBFNN)

Citation - S Pazhanirajan, P Dhanalakshmi. "Classification of Parkinsons Disease using MRI Images." International Journal of Computer Science and Software Engineering 5, no. 10 (2016): 233-237.